HIV-1 uses frequent recombination between its two RNA genomes to create viral diversity. This diversity helps the virus to escape host immune response and drug therapy. Recombination can occur by strand transfer between RNA templates. This mechanism is also employed for the minus strand strong stop transfer in the replication pathway. Transfer involves a shift of the growing cDNA primer from the original donor RNA to a second acceptor RNA. Our work reconstituting recombination in vitro with pure proteins, and in vivo in cell culture, addresses mechanisms that drive strand transfer. Transfers initiate at sites where the virally encoded reverse transcriptase (RT) pauses, allowing it to use its RNase H function to concentrate cuts in the donor template. Transfers can occur by a multi-step process in which acceptor template invades the DNA at the gapped site in the donor template. The cDNA-acceptor hybrid spreads until the 3'terminal region of the cDNA completes transfer. However, major parts of the transfer mechanism are unexplored. Minus strand transfer model reactions in vitro indicate that RNA and DNA folding is an important determinant of transfer efficiency. We are investigating evidence that folding contributes to a time-dependent inactivation of cDNA ends for transfer, and that high efficiency depends on mechanisms that complete transfer before inactivation can occur. New results indicate that transfers can occur by a mechanism called proximity that does not involve spreading of the initial hybrid. We will evaluate the relative contributions of the spreading versus proximity mechanisms. Evidence suggests that the RT is obligated to dissociate for transfers, and that the RT must exercise its unique 5'end-directed RNase H activity. We are determining whether either or both functions are essential for transfer. Lastly, we developed a viral cell culture system that measures the positions and frequencies of recombination crossovers over more than half of the length of HIV-1 at a resolution of 25 nucleotides. We initially sequenced a 459 bp region from DIS through part of the gag gene. It revealed a striking peak of recombination in which two-thirds of crossovers in the region occurred within about 100 nucleotides. Significantly, we successfully recapitulated the hot spot in strand transfer assays in vitro, allowing us to determine its structural and mechanistic basis. Overall, results of our work will clarify the exact mechanisms and requirements of strand transfer in HIV-1. This is a first step to therapeutic targeting of strand transfer as a means of interfering with HIV-1 infection.PUBLIC HEALTH RELEVANCE: HIV-1 rapidly evolves its structure in infected people. This allows it to escape immune response and to develop resistance to all current attempts at drug therapy. The virus has a mechanism whereby it can combine different drug resistance traits that it inherited from two different virus parents. This mechanism can produce viruses with increased or multi-drug resistance. Our results will help us to understand and defeat this mechanism so that anti-AIDS drugs can be more effective.